Adaptive LIDAR FoV for Autonomous Vehicle Elevation Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional LIDAR sensor units in autonomous vehicles have an arbitrarily determined Field of View (FoV) that is not adaptable to different terrains, leading to the potential misclassification of road features as obstacles and failure to detect objects below the lower viewing angle.

Innovation Solution

An elevation detection system comprising a vertically movable LIDAR sensor unit and a computation unit that adjusts the FoV by determining a lower limit value based on the lowest elevation of objects and road boundary elevation, allowing the LIDAR sensor unit to disregard objects below this threshold as obstacles.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the LIDAR sensor unit uses a fixed Field of View (FoV) to detect surrounding objects, then the device structure is simple and easy to manufacture, but it cannot adapt to different terrain conditions and may misclassify road features as obstacles

Engineering Contradiction:

Improveadaptability to different terrainVSAvoidsensor system complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a dynamically adjustable FoV in the LIDAR sensor unit, allowing the lower viewing angle to be modified based on detected road boundary elevation. This enables the sensor to adapt to varying terrain conditions (uphill, downhill, flat roads) while maintaining a relatively simple overall device structure. The computation unit calculates the appropriate FoV adjustment based on elevation data, making the system dynamic without requiring complex mechanical reconfiguration.

Inventive Principle:

Principle #15Dynamics

2Measurement precision

If the LIDAR sensor unit detects all low elevation points, then measurement precision is improved, but the road may be misclassified as an obstacle causing navigation errors

Engineering Contradiction:

Improveelevation detection precisionVSAvoidobstacle classification accuracy

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent applies preliminary action by first detecting the road boundary elevation using the elevation sensor unit before the LIDAR sensor unit processes obstacle detection. The computation unit uses this preliminary elevation information to establish a reference level, allowing subsequent LIDAR data to be interpreted in context. This preliminary detection of road geometry prevents misclassification of low-elevation road features as obstacles while maintaining precise elevation measurement.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by continuously monitoring detected elevation data and adjusting the lower FoV limit accordingly. The computation unit analyzes the relationship between detected object elevations and road boundary elevation, then provides feedback to the LIDAR sensor unit to adjust its filtering criteria. This closed-loop feedback mechanism ensures accurate distinction between road surface and actual obstacles while maintaining high measurement precision.

Inventive Principle:

Principle #23Feedback

3Productivity

If the LIDAR sensor unit ignores objects below a certain elevation threshold, then productivity is improved by reducing data processing, but objects below the threshold may be missed including both road features and actual obstacles

Engineering Contradiction:

Improvedata processing efficiencyVSAvoidobstacle detection reliability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent dynamically changes the elevation threshold parameter based on detected road boundary conditions. Instead of using a fixed threshold, the computation unit calculates the lower FoV limit as a variable parameter that adapts to the specific terrain context. This parameter change enables efficient data processing by filtering out road surface points while maintaining reliability by ensuring actual obstacles above the adaptive threshold are not missed.

Inventive Principle:

Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables accurate detection of road boundaries and obstacles, improving navigation efficiency by reducing unnecessary data processing and adapting to varying terrain conditions.

Implementation Method 1

Light Detection and Ranging (LIDAR) sensor unit is configured to detect the surrounding of the AV

Methodology Applied

Scientific EffectLIDAR: LIDAR

Implementation Method 2

The elevation sensor unit is configured to detect an elevation of a plurality of objects having a lower most elevation, in the surrounding of the AV

Methodology Applied

Scientific EffectElevation detection:

Data Source

PatentUS11536846B2Elevation detection system for autonomous vehicles and a method thereof

Publication Date: 2022.12.27 WIPRO LTD

AI summary

The present disclosure relates to an elevation detection system for an Autonomous Vehicle (AV) and a method for detecting elevation of surrounding of the AV. The elevation detection system includes an elevation sensor unit and a computation unit. The elevation sensor unit is configured to detect an elevation of a plurality of objects having a lower most elevation, in the surrounding of the AV to determine a boundary elevation of the road. The elevation sensor unit is vertically movable within a range of vertical positions. A Light Detection and Ranging (LIDAR) sensor unit is associated with the elevation sensor unit, to detect the surrounding of the AV, having a predefined Field of View (FoV). The computation unit determines a lower limit value of the FoV and provides it to the LIDAR sensor unit for accurately detecting obstacles in the road.